Synthesis and evaluation of analogues of congo red as potential compounds against transmissible spongiform encephalopathies

Article abstract of DOI:10.1016/S0223-5234(03)00081-3

The synthesis of analogues of the amyloid stain Congo red (1) as potential compounds against transmissible spongiform encephalopathies (TSEs) is reported. Using the direct method, aniline (2) or diamines such as 4,4′-diaminodiphenylsulfone (dapsone, 9), 3,3′-diaminodiphenylsulfone (10), benzidine (11), 3,3′-dimethoxybenzidine (12) or 3,3′-dichlorobenzidine (13) were diazotised to afford the corresponding diazonium salts, which without isolation, were directly used for coupling with a range of aromatic sulfonic or carboxylic acids to provide the corresponding truncated dyes analogues of Congo red, 4, 6, 8, and the symmetrical bis azoic dyes 14-19, 21-22, 24 and 26-29 as their sodium salts. Compounds were assayed in a cellular model of scrapie, a sheep TSE. Some of the compounds were shown to have similar activity to the lead compound Congo red. Molecular modelling was carried out to investigate potential structure-activity relationships (SARs) relating to the size and shape of Congo Red analogues. Within the range of compounds tested no discernible SARs were found.

Full text of DOI:10.1016/S0223-5234(03)00081-3

European Journal of Medicinal Chemistry 38 (2003) 567ꢀ579
/
www.elsevier.com/locate/ejmech
Orginal article
Synthesis and evaluation of analogues of congo red as potential
compounds against transmissible spongiform encephalopathies
a
a
b
b
H e´ l e` ne Rudyk , Michael H. Knaggs , Snezana Vasiljevic , James Hope ,
b
Chris Birkett , Ian H. Gilbert *
a,
a
Welsh School of Pharmacy, Cardiff University, Redwood Building, King Edward VII Avenue, Cardiff CF10 3XF, UK
b
Institute of Animal Health, Compton Laboratories, Compton, Newbury, Berkshire RG20 7NN, UK
Received 22 October 2002; received in revised form 13 March 2003; accepted 17 March 2003
Abstract
The synthesis of analogues of the amyloid stain Congo red (1) as potential compounds against transmissible spongiform
encephalopathies (TSEs) is reported. Using the direct method, aniline (2) or diamines such as 4,4?-diaminodiphenylsulfone (dapsone,
9), 3,3?-diaminodiphenylsulfone (10), benzidine (11), 3,3?-dimethoxybenzidine (12) or 3,3?-dichlorobenzidine (13) were diazotised to
afford the corresponding diazonium salts, which without isolation, were directly used for coupling with a range of aromatic sulfonic
or carboxylic acids to provide the corresponding truncated dyes analogues of Congo red, 4, 6, 8, and the symmetrical bis azoic dyes
1
compounds were shown to have similar activity to the lead compound Congo red. Molecular modelling was carried out to
4ꢀ
/
19, 21ꢀ
/
22, 24 and 26ꢀ29 as their sodium salts. Compounds were assayed in a cellular model of scrapie, a sheep TSE. Some of the
/
investigate potential structure-activity relationships (SARs) relating to the size and shape of Congo Red analogues. Within the range
of compounds tested no discernible SARs were found.
´
#
2003 Editions scientifiques et m e´ dicales Elsevier SAS. All rights reserved.
Keywords: Prions; Azo dyes; TSE
1
. Introduction
Transmission of disease within a single species occurs
much more readily than inter-species transmission. The
infectious agent either requires or is an abnormally
Transmissible spongiform encephalopathies (TSEs or
prion diseases) are rare progressive diseases giving rise
to degeneration of the brain and are fatal with no
c
folded prion protein. The normal prion protein (PrP ),
occurs in many tissue types, but especially in the central
nervous system on the outer surface of neurons and glial
cells. The exact function of PrP is nonetheless poorly
understood [5,6]. The lack of prion protein has been
reported to change the circadian activity rhythms [7] and
sleep in mice. The prion protein has an effect on the
intracellular free calcium levels [8], and has also been
linked to super-oxide dismutase activity [5].
known method of treatment [1ꢀ3]. The diseases can be
/
sporadic and are naturally occurring in some organisms
and are linked more or less strongly to genetic muta-
tions. TSEs can also be caused by infection. Thus
inoculation of the infectious agent into the brain can
give rise to disease. Examples of TSEs include scrapie in
sheep; chronic waste disease in elk and deer; bovine
spongiform encephalopathy in cattle; fatal familial
insomnia, Creutzfeld Jakob disease and Kuru in humans
c
In spongiform encephalopathies, the PrP undergoes a
conformational change leading to the abnormal form or
[4]. The diseases exhibit a long asymptomatic incubation
period, leading inevitably to a fatal neurological illness.
res
abnormal prion protein (PrP ) [9]. The abnormal
res
isoform of the prion protein (PrP ) differs from the
c
normal cellular form (PrP ) in that it has a much higher
resistance to proteinase K and has much lower solubi-
c
lity. The normal cellular protein PrP has been shown to
* Corresponding author.
E-mail address: gilbertih@cf.ac.uk (I.H. Gilbert).
´
0
223-5234/03/$ - see front matter # 2003 Editions scientifiques et m e´ dicales Elsevier SAS. All rights reserved.
doi:10.1016/S0223-5234(03)00081-3

568
H. Rudyk et al. / European Journal of Medicinal Chemistry 38 (2003) 567ꢀ579
/
have a a-helical structure (42%) with little b-sheets (3%).
published by Demaimay et al. [23], which discusses
some structure-activity relationship (SAR) studies on
Congo red. This present study extends their work and
presents additional results and conclusions. We thus
were interested in the synthesis of analogues of Congo
red (1) and several modifications are reported. The
synthesis of azoic compounds is widely described [24]
and the first step of the synthesis involved the formation
of the diazonium salt then coupling with aromatic
compounds. The procedures used depend on the nature
of the arylamine. We mainly used the direct method,
which consisted in adding a solution of a metallic nitrite
to a cold solution of the arylamine in aqueous mineral
acid (the arylamine may be partly in suspension in the
form of a salt). We studied then the coupling of several
diazonium salts, obtained from either monoamines or
diamines, with a range of sulfonic or carboxylic acids
derivatives.
Congo red (1) is a symmetric bis-azo dye and there-
fore we decided to synthesise firstly the truncated
molecules (4), (6) [25] and (8) (Scheme 1) corresponding
to only half of the molecule. These compounds provided
some information regarding the necessity of the whole
molecule for activity, as well as the importance of
features such as the nature of the aromatic core (from
naphthyl ring to the phenyl ring) or the possibilities of
substitution of the sulfonic acid by the corresponding
carboxylic derivative. The diazonium salt of aniline (2)
was obtained according to literature [26] and coupled,
without isolation of the diazonium salt, to either
sulfanilic acid (3), 4-amino-naphthalenesulfonic acid
res
In contrast the PrP has a high b-sheet (43%) content
and a slightly reduced a-helical structure (30%) [10].
Both isoforms of PrP appear to have the same amino
acid sequence, and the same covalent modifications,
including a single disulfide bond, two complex-type N-
linked oligosaccharide chains and a C-terminal glyco-
phosphatidyl-inositol (GPI) anchor [11]. During the
res
course of TSEs, the PrP forms amyloid plaques. It is
uncertain if these plaques themselves have a neurotoxic
effect or are a result of the disease.
Several classes of compounds including azo dyes,
sulphated polysaccharides, porphyrins and polyene
antibiotics [12ꢀ15] have been shown to have activity
/
against models of prion diseases [16] and on models of
other amyloid diseases such as Alzheimer’s disease
[
12,14]. We were particularly interested by the azo
dyes, as these may represent small molecules, which
may be amenable to structure activity studies. In
particular Congo red (1, Fig. 1) is an important lead
molecule. It has been shown to ‘cure’ persistently
scrapie-infected cells in cell culture. The amyloid-bind-
ing dye Congo red has been reported to prolong the
incubation period in scrapie infected hamsters [17] when
inoculated intracerebrally, presenting modest therapeu-
tic anti-scrapie activity in vivo. The effect of Congo red
is similar to the one observed with sulphated glycans
[
res
13,15], as it prevents new accumulation of PrP ;
furthermore 1 also inhibits the replication of scrapie
infectivity.
(
5), or 4-aminobenzoic acid (7), commercially available
as either their free acids or as their sodium salts. At
completion of coupling, as indicated by analysing a
sample by mass spectroscopy, addition of 20% sodium
chloride solution to the reaction mixture afforded after
filtration the corresponding dyes as their sodium salts
together with inorganic salts. Purification was obtained
either by preparative TLC (1-butanol:ethanol:ammo-
nium hydroxide:pyridine / 4:1:2:3) [27] to yield the dyes
2
. Chemistry
Although Congo red (1) shows anti-TSE activity, it is
not a good drug candidate because of non-specific
binding, toxicity [18ꢀ20] and poor penetration of the
bloodꢀbrain barrier. The inherent carcinogenicity of 1 is
/
/
probably due to the degradation into benzidine [21] (11)
in the digestive system and further accumulation in the
bladder, leading to bladder cancer [22] if large doses are
ingested. We proposed to carry out structure-activity
studies in order to identify important features of the dye
for activity, to circumvent these problems (selectivity,
membrane permeability) and to increase activity. Dur-
ing the course of this present work, a paper was
(
4), (6) and (8) as their pure sodium salts or ion
exchange chromatography (Dowex 50WX8-200) in the
case of 4 and 6 to give the free acids.
As the anti-TSE activity of such truncated molecules
turned out to be significantly decreased [28] (Table 1) in
comparison to 1, indicating the importance of the whole
molecule, we decided to synthesise some close analogues
of Congo red (1) by changing either the nature of the
biphenyl spacer originally observed in 1 (length, pre-
sence of substituents) or the nature of the acid.
2.1. Variation of spacer
Commercially available diamines (Fig. 2), 4,4?-dia-
mino-diphenylsulfone (dapsone, 9), 3,3?-diamino-diphe-
nylsulfone (10), benzidine (11), 3,3?-dimethoxybenzidine
Fig. 1. Congo red.

H. Rudyk et al. / European Journal of Medicinal Chemistry 38 (2003) 567ꢀ
/
579
569
Scheme 1. (a) HCl 10%, 0 8C, NaNO
2
1.05 equiv.; (b) sulfanilic acid (3), Na
2
CO
3
, 0 8C0r.t.; (c) NaCl; (d) 4-amino-naphthalenesulfonic acid (5),
/
Na
2
CO
3
, 0 8C0/r.t.; (e) 4-amino-benzoic acid (7), Na
2
CO
3
, 0 8C0r.t.
/
(
12) and 3,3?-dichlorobenzidine (13) were selected. This
should give some variation in the length and also in the
electronic and steric properties of the linker. In addition
the sulfones 9 and 10 should give rise to different
dihedral angles between the two aromatic rings of the
linker, alter the electronic properties of the linker and on
metabolism give rise to less toxic breakdown products.
The 3,3?-dimethoxybenzidine (12) and 3,3?-dichloroben-
zidine (13) were selected as they should cause minor
alterations to the steric and electronic properties of the
linker.
Fig. 2. Linkers.
These linkers were diazotised according to literature
precedent [24,26] and coupled with either sulfanilic acid
(
3) or 4-amino-naphthalenesulfonic acid (5) to give
Table 1
Dihedral angle between phenyl groups and distance between acid residues
res
PrP (% of controls)
˚
Distance acid in A
Compd. Number as in Ref. [28]
Angle
1
00 mM
1 mM
Method 1
20.34
Method 2
Method 3
Method 1
40.1
Method 2
46.6
Method 3
57
a
1
4
6
8
1
1
1
1
1
1
2
2
2
2
CR
IV
5.5
53.3
99.5
21.07
20.24
29.2
63
III
129.2
99.1
86.7
V
XI
26.7
62.7
8.5
b
c
c
c
4
5
6
7
8
9
1
2
4
6
19.75
17.49
8.31
21.35
20.60
12.71
20.99
21.07
20.29
20.46
21.40
14.91
20.42
8.60
5.68
b
IX
X
117.3
142.8
184.6
218
b
23.7
5.7
9.10
20.43
a
VII
VI
20.08
20.04
20.36
20.45
20.30
14.86
21.46
40.7
41.8
40.6
43.6
41.7
43.3
47.8
47.1
46.4
46.9
45.9
48.4
55.2
55.6
56.7
55.2
56.8
57.2
a
2.9
4.6
20.94
20.18
a
XII
XIV
XIII
XV
XXI
187.1
148.5
223.7
218.3
103.3
a
10.5
3.6
20.28
20.24
a
5.4
46.8
14.10
a,d
8.9
c
1
8.2
7.27
a,d
27
XXII
56.4
72.7
16.63
18.71
c
1
8.99
2
8
9
XIX
XX
2.5
23.5
107.2
115.4
20.20
20.15
20.20
20.44
19.84
19.79
39.7
43.5
47.2
48.3
54.9
54.7
2
Method 1: Monte-Carlo conformational search. Method 2: quantum mechanics. Method 3: simulated annealing.
a
H-bonding between either the amino or the hydroxyl and the azo bonds.
Population of conformers with strong intramolecular H-bonding between the amino or the hydroxyl and the sulfonic group on the other naphtyl
b
or phenyl unit.
c
Sulfones angle not defined.
d
Second population of conformers within a range of 3ꢀ
/
5 kcal from the population of conformers of lowest energy. The azo bonds present a
staggered conformation.

570
H. Rudyk et al. / European Journal of Medicinal Chemistry 38 (2003) 567ꢀ579
/
Scheme 2. (a) 4,4?-Diaminodiphenyl sulfone 9, HCl, NaNO
2
, 0 8C; (b) Na
2
CO
3
, 0 8C0r.t., then reflux 1 h; (c) 10% NaCl; (d) linkers 9, 10, 12 or 13,
/
HCl, NaNO , 0 8C; (e) Na CO , 0 8C0r.t. then 50 8C for 12 and 13.
2
2
3
/
compounds 14ꢀ
/
18 (Scheme 2). The products were
50 8C. This temperature was sufficient to obtain the dyes
with good yield after 4ꢀ5 h.
contaminated with sodium chloride and then purified
by soxhlet extraction followed by preparative TLC. The
structure of each compound was determined by one and
/
2.2. Variation of the terminal substituents
1
13
or two dimensional H and C-NMR experiments.
Compound 15 [29] was prepared according to the
general procedure. The procedure for the preparation
of compound 14 was slightly modified as the reaction
did not go to completion when sulfanilic acid (3) was
coupled to the diazonium salt of 4,4?-diaminodiphenyl-
sulfone (9). Compound 14 was prepared as a bright
yellow solid after a 1 h reflux of the aqueous solution of
the reaction mixture followed by cooling and addition of
Systematic variation of the terminal substituents was
undertaken. Firstly the necessity of the amino function-
ality was probed, and then variation of the sulfonic acid
was carried out.
2
.2.1. Variation of the amino functionality
The first variation attempted was to acetylate the
amino functionality in Congo red. Reacting Congo red
with acetic anhydride did not yield the desired N-
acetylated derivative. However, reaction with trifluor-
oacetic anhydride led to conversion to the trifluoroace-
tamide derivative (19) (Scheme 3). Thus a solution of
Congo red (1) in anhydrous N,N’-dimethylformamide
was treated with an excess of trifluoroacetic anhydride
at 0 8C for 1 h to give the N-protected compound. The
solvents were removed and the product washed several
1
0% sodium chloride solution. The dye was obtained
with a low yield, probably resulting from the degrada-
tion of the diazonium salt through the heating. The
formation of compounds 17 [30] and 18 [31] resulting
from the coupling of 4-amino-naphthalenesulfonic acid
(
(
5) with the diazonium salts of 3,3?-dimethoxybenzidine
12) and 3,3?-dichlorobenzidine (13) required also
further heating of the reaction mixture. In order to
prevent decomposition of the diazonium salt, we
decided not to heat the aqueous solution higher than
1
9
times with methanol to give 19, as indicated by F-
NMR experiment and mass spectroscopy.
Scheme 3. (a) (CF
3
CO)
2
O, DMF, 0 8C0r.t., 1 h.
/

H. Rudyk et al. / European Journal of Medicinal Chemistry 38 (2003) 567ꢀ
/
579
571
Scheme 4. (a) Linkers 9, 11 or 12, HCl, NaNO
Na CO r.t., then 50 8C.
0 8C0
2
, 0 8C; (b) Na
2
CO
3
0 8C0
/
r.t., then 50 8C;(c) 10% NaCl. (d) Linker 12, HCl, NaNO
2
, 0 8C; (e)
2
3
/
A series of compounds were also prepared in which
the amino functionality was replaced by a hydroxyl
group (Scheme 4). Benzidine (11) and 3,3?-dimethox-
ybenzidine (12) were diazotised according to literature
carboxylate groups as with Chrysamine G (28) leads to
compounds with improved binding to b-amyloid [12,36]
and better pharmacokinetic properties. We thus studied
the coupling (Scheme 5) of the different diamines, 4,4?-
diaminodiphenylsulfone (9), 3,3?-diaminodiphenylsul-
fone (10), benzidine (11) and 3,3?-dimethoxybenzidine
(12) with salicylic acid (25). Chrysamine G (28) was
previously synthesised [37,38] according to a similar
procedure. All reactions proceeded at room temperature
[
24] and the corresponding diazonium salts were coupled
at 0 8C with 4-hydroxynaphthalene sulfonic acid (20) to
give diazo compounds 21 [32] and 22 [33]. As previously
discussed those were contaminated with sodium chlo-
ride. The structure of each compound was determined
1
13
by one and or two dimensional H and C-NMR
experiments. The synthesis of compounds 21 and 22
required further heating of the aqueous solution at
(r.t.), once the diazonium solution was added at 0 8C.
The evolution of the coupling was monitored by mass
spectroscopy. Addition of the solution of the diazonium
5
0 8C in order to obtain the corresponding dyes. This
salts of 9ꢀ/13 obtained according to literature [24] to a
temperature was sufficient to permit the completion of
the reaction and prevent degradation of the diazonium
salt.
cold solution of 25 in Na CO was followed after
2
3
completion of coupling by addition of 10% NaCl
solution to give the diazo compounds 26ꢀ29 [39] as
/
One further variation was attempted by coupling 6-
amino-1-naphthol-3-sulfonic acid (23) (Scheme 4) to the
diazonium salt of 3,3?-dimethoxybenzidine (12) which
should give further variation in the terminal region. The
orientation of the coupling can be controlled as the
coupling is reported to be pH dependent [34,35] this
controlled reaction was used in the manufacture of
direct dyes to obtain bis-azo dyes. During this coupling
dark brown solids, which were then purified by soxhlet
extraction using 95% ethanol (compounds 26, 27, 29) or
preparative TLC using 1-butanol:ethanol:ammonium
hydroxide:pyridine / 4:1:2:3 for compound 28 [27].
reaction, the pH was controlled at ꢀ
/
7ꢀ
/
8 as at pH B
/
3ꢀ
/
4
the coupling occurred on the ring bearing the amino
functionality. In basic conditions, the coupling occur in
ortho of the hydroxyl moiety giving rise to the desired
compound. No starting material 23 was detected by
mass spectroscopy after completion of the reaction and
1
13
the experimental H and C-NMR of the product 24
are in agreement with the proposed structure.
2
.2.2. Variation in the nature of the acid
The replacement of the sulfonate group with less
acidic groups, which should improve pharmacokinetic
properties and selectivity, was then studied. This has
precedence from the work on Alzheimer’s disease where
it has been shown that replacing the sulfonate with
Scheme 5. (a) Linkers 9, 10, 11 or 12, HCl, NaNO
2 2 3
, 0 8C; Na CO ; (b)
Na CO r.t.; (c) 10% NaCl.
2
3
0 8C0
/

572
H. Rudyk et al. / European Journal of Medicinal Chemistry 38 (2003) 567ꢀ579
/
3
. Modelling studies
3.3. Method 3: simulated annealing
Studies in models of Alzheimer’s disease [12] and
Another set of results was obtained by loading in
Sybyl the minimised structures previously obtained in
Mopac followed by simulated annealing (using
Demaimay’s study [23] of prion diseases have indicated
the importance of the dihedral angle between the two
phenyl rings of the benzidine for activity. In these
studies, active compounds were calculated to have a
dihedral angle of about 358. When the dihedral angle
was significantly increased (for example by 2,2?-substi-
tution) activity decreased.
In this study, three methods were investigated to
calculate the dihedral angle. In addition, we decided to
measure the distance between the acidic groups to see if
this led to any SARs. There are several difficulties that
need to be taken into account with the modelling
studies:
GeisteigerꢀH u¨ ckel charges, the tripos forcefield and
/
distance dependent dielectric constant to help mimic the
shield effects of solvent) for each compound by heating
at 1000 K during 1000 fs and then cooling to 0 K for
2000 fs. This cycle was repeated for 100 times. We
selected then for each compound the conformations of
lowest energy obtained. The predicted dihedral values
are reported in Table 1 and generally are larger by 15ꢀ
/
178. Interestingly strong intramolecular hydrogen bond-
ing was observed (Table 1). Insertion of a sulfone
(compounds 14, 15 and 16) within the biphenyl system
induced strong intramolecular hydrogen bonding be-
tween one part of the molecule, bearing the sulfonic
moiety, with the amino moiety of the other part, thus
forcing the molecule to loose its symmetry and then
adopting a ‘cage-like’ conformation (Fig. 4). The set of
results obtained for compounds 14 and 15 bearing a
sulfone in position 4,4?- is different using this last
method in comparison to the previous data. This may
be due to insufficient searching of conformational space
in methods 1 and 2, when compared to simulated
annealing.
a) The ‘active’ conformation, presumably bound to the
c
abnormal or PrP , may not be in the lowest energy.
b) The molecules will be solvated, which will have an
effect on the conformation.
c) There may be several low energy conformations.
d) The dyes may exist as a supramolecule [40ꢀ42] in
/
solution depending on the concentration of dye.
3
.1. Method 1: Monte-Carlo conformational search
We carried out molecular modelling on O2 Silicon
Graphics workstations for each compound using
4. Biological data
†
Macromodel
6.0 and performing a Monte-Carlo
conformational search with the Amber force field and
the GB/SA solvation model for water as implemented in
Macromodel. The structures were drawn in the ‘draw’
mode with the double bond trans and the sulfonate and
carboxylate groups represented in the anionic form. The
dihedral angle and the intra-molecular distance between
either the two sulfonates or the two carboxylate groups
of each dye were measured and the set of results
obtained for each compound is shown in Table 1.
The compounds synthesised were assayed for their
c
ability to inhibit the conversion of PrP to its proteinase-
res
resistant conformer (PrP ) in scrapie-infected SMB
cells. Compounds were tested at two different concen-
trations, 100 and 1 mM, using Congo red (1) as
reference. This evaluation has been published in detail
elsewhere [28] but the results are presented in Fig. 3 to
facilitate cross-referencing and discussion of structure-
biological activity relationships.
5. Discussion
3
.2. Method 2: quantum mechanics
The aim of this work was the synthesis and evaluation
Similar data was obtained using quantum mechanics
calculations (in vacuum) by minimising the compounds
in Sybyl (using GeisteigerꢀH u¨ ckel charges, the tripos
of some analogues of Congo red (1) as potential anti-
scrapie agents with persistently scrapie-infected SMB
cells as a model for TSEs. Systematic variation was
carried out on a number of different features of the
molecule.
/
forcefield, distance dependent dielectric constant to help
mimic the shield effects of solvent and steepest descent
minimization utilising an energy gradient cut-off at
ꢁ
1
0
.001 kcal mol ) and subsequent loading of the
minimised structures in Mopac (PM3 hamiltonian with
full geometry optimization in internal coordinate space).
5.1. The symmetry
This predicted dihedral values larger by about 5ꢀ
similar bond length values.
/
68 and
Congo red is a symmetrical molecule. Our data
suggests that the whole molecule is required for full

H. Rudyk et al. / European Journal of Medicinal Chemistry 38 (2003) 567ꢀ
/
579
573
Fig. 3. Biological activity.
activity (compare compounds 4, 6 and 8 with compound
1
having an H-bond donor at this position cannot be ruled
out.
).
5
.2. The spacer
5.4. The acid group
It is possible to make some changes in the central
spacer without a major effect on activity at 100 mM,
though there is some loss of activity at 1 mM (com-
Congo red (1) has a sulfonic acid group. At physio-
logical pH this group is completely ionised and will
probably prevent uptake across the bloodꢀbrain barrier
12]. Compounds 26ꢀ29 represent compounds in which
the sulfonate has been replaced by a carboxylate, which
should have greater permeability across the bloodꢀbrain
barrier. With the benzidine-derived compounds 28ꢀ29,
there was little effect on activity compared to Congo red
/
pounds 14ꢀ18). Thus compounds 14, 15 and 16 have
/
[
/
sulfones in the molecule. Compounds 17 and 18 have
substituents on the 3,3?-positions of the benzidine ring.
These do not seem to have a large effect on the dihedral
angle of the molecule and hence not on the conforma-
tion of the molecule.
/
/
(1) at 100mM. However with compounds 26 and 27, the
sulfone analogues, there was lower activity.
5
.3. The amino group
Modelling was undertaken with three different meth-
ods to see if there were any major differences. The
features that have been mentioned as possibly being
important for activity are the distance between the
sulfonic acid moieties and the dihedral angle across
the benzidine unit [23]. For most of the compounds
described here, the three modelling methods predicted
relatively similar distances and dihedral angles (where
these could be determined) (Table 1).
Acetylation of the amino groups with trifluoroacetic
anhydride (compound 19) did not have a large effect on
activity at 100mM. Similarly replacement of the amino
group with a hydroxyl group (compounds 21, 22 and 24)
did not seem to have a marked effect on activity at 100
mM. This means that it is not necessary to have a
positive charge at this position, although the necessity of

574
H. Rudyk et al. / European Journal of Medicinal Chemistry 38 (2003) 567ꢀ579
/
5
.5. Benzidine derived analogues (compounds 1, 17 and
8)
azo bond (Fig. 4). This decrease in activity could be due
to the coexistence of those two populations. When
1
compounds 14ꢀ16 were studied by simulated annealing,
/
In the examples described here there did not seem to
the lowest energy conformations predicted strong local
H-hydrogen bonding (Table 1, Fig. 4) which brought the
two sulfonic acid groups closer to each other, as well as
intramolecular p-stacking. It is interesting that some of
the sulfones have activity despite having a bend in the
molecule. This bend could reduce the possibility of
intermolecular p-stacking, which has been speculated to
be important in the mechanism of these types of
be a very wide variation in distance and angle and there
was no discernible relationships between activity and
either distance or dihedral angle. Demaimay et al. [23]
reported some compounds (2,2?-substituted benzidines)
which had a significantly larger dihedral angle, than
those 3,3?-substituted molecules; these compounds ap-
peared to have lower activity.
compounds [40ꢀ42].
/
5
.6. Sulfone analogues (14ꢀ
/
16 and 26ꢀ
/
27)
Therefore in summary, a number of azo dyes related
to Congo red have been prepared and then screened
against a persistently scrapie-infected cell line. It proved
possible to make modifications to the structure of
Congo red to either increase its lipophilicity or to
remove the central benzidine unit without having a
large reduction in activity against the scrapie model of
infection. According to the modelling, most of the
compounds contained similar dihedral angles at the
biphenyl ring and a relatively similar distance between
the charged sulfonic acid/carboxylic acid groups.
An interesting variation described here is the sulfones.
Two types of sulfone analogues were synthesised with
the sulfone either in position 1,1?- (14, 15 and 26) or in
position 2,2?- (16 and 27). The sulfones introduce a bend
into the molecule, destroying the planarity of the linker
unit. A small decrease in activity was observed at 100
mM when modifying the position of the sulfone from
1
,1?- (15) to 2,2?-(16). Compound 15 had thus similar
activities to the benzidine derived analogues (1, 17 and
1
2
8). The exceptions to this were compounds 14, 26 and
7. The differences in activity observed for 14 may be
related to the modification of the aromatic ring from a
naphthyl unit as in 1 to a phenyl unit, rather than the
presence of a sulfone. Compounds 26 and 27 showed
decreased activity at 100 mM compared to their benzi-
dine analogues 28 and 29. Interestingly, two populations
6. Experimental part
All solvents and chemicals were purchased either from
Fluka, Aldrich or Jansen. Water refers to deionised
water. Preparative TLC were carried out either on pre-
coated silica plates (Kieselgel 60 F254 BDH) or on pre-
ꢁ
1
of conformers were observed (within 3 kcal mol
)
when using simulated annealing. The population of
conformers of lowest energy presented strong intramo-
lecular H-bonding between the hydroxyl moiety and the
coated silica plates made from silica (Silica gel GF254
with visualisation via UV light and/or KMnO solution.
)
4
1
All H, C and F-NMR spectra were recorded on a
13
19
Fig. 4. Lowest energy conformations from simulated annealing (method 3); (A) congo red, compound 1; (B) compound 15; (C) compound 27.

H. Rudyk et al. / European Journal of Medicinal Chemistry 38 (2003) 567ꢀ
/
579
575
Bruker Advance DPX300 spectrometer at 300, 75 and
82.38 MHz, respectively. Chemical shifts are reported
6.2. 4-Amino-3-(2-phenyl-1-diazenyl)-1-
naphthalenesulfonic acid, sodium salt (6) [25]
2
downfield in parts per million using CDCl3 as an
internal reference at 7.27 ppm unless otherwise stated
and all coupling constants (J values) are in hertz. Low-
resolution mass spectra were recorded on a Fisons VG
platform Electrospray mass spectrometer. High-resolu-
tion mass spectra were recorded by the EPSRC National
Mass Spectrometry Service Centre (Chemistry depart-
ment, Swansea). The name of the different compounds
was given according to the IUPAC rules and we decided
to number the compounds as indicated on Fig. 5 for the
A solution of the diazonium salt of aniline (52 mL)
was added dropwise at 0 8C to a solution of 4-amino-
naphthalenesulfonic acid sodium salt (as its bis-hydrate)
(
5) (2.63 g; 10.97 mmol) in 5% Na CO (60 mL). The
2 3
red-orange residue obtained after filtration was dried to
give the dye as its sodium salt with inorganic salts. A
sample (200 mg) was treated by soxhlet extraction (200
mL; 95% ethanol) to give 157 mg of a yellow dark solid.
95 mg of compound 6 were also purified directly by
preparative TLC (1-butanol:ethanol:ammonium hydro-
xide:pyridine / 4:1:2:3) to yield 71 mg as the pure
compound. Treatment of another sample by ion ex-
change resin (Dowex 50WX8-200) gave the correspond-
1
13
purpose of simplification of the H and C-NMR
spectra. Molecular modelling was carried out on O2
Silicon Graphics workstations. Appropriate safety mea-
sures were used for compounds 11, 12 and 13, which are
known to be carcinogenic [21].
1
ing free acid. H-NMR (300 MHz, CD OD) d 8.78 (d,
3
H
1
H, Jꢂ
Jꢂ8.2 Hz, H-11), 7.89 (AB, 2H, H-3, H-5), 7.65 (dd,
H, Jꢂ1.2 and 8.3 Hz, H-13), 7.65ꢀ7.48 (m, 3H, H-1,
H-2, H-6), 7.41 (d, 1H, Jꢂ1.2 Hz, H-12); C-NMR (75
MHz, CD OD) dC 151.63 (C-4), 145.46 (C-10), 131.56
/
8.3 Hz, H-14), 8.62 (s, 1H, H-8), 8.31 (d, 1H,
/
1
/
/
1
3
/
6.1. 4-Amino-3- (2-phenyl-1-diazenyl)-1-benzenesulfonic
acid, sodium salt (4)
3
(C-9), 129.26 (C-14a), 129.22 (C-7), 125.09 (C-10a),
1
1
22.23 (C-8); 129.83, 129.33, 128.99, 127.49, 125.09,
ꢁ
Aniline (2) was diazotised by adding 1 equiv. of 4 N
NaNO (5.5 mL) to a solution of the amine (2 mL; 21.47
23.65, 123.07 (ring); HR-ES -MS Calc. for
ꢃ ꢁ
2
C H N O SNa [M-Na ] 326.0599 Found: 326.0600.
3
1
6
12
3
mmol) in a 2% HCl solution (100 mL) at 0ꢀ
/
5 8C. After
completion of diazotisation (ꢁ2 h), the light orange
/
6.3. 4-Amino-3-(2-phenyl-1-diazenyl)-1-benzoic acid,
sodium salt (8)
diazonium solution was divided in two equal parts and
directly used. A solution of the diazonium salt (52 mL)
was added dropwise to a solution of sulfanilic acid (3)
Aniline (2) was diazotised at 0 8C by addition of a 4 N
(
1.869 g; 10.97 mmol) in 5% Na CO (60 mL). The light
2 3
NaNO solution (2.75 mL; 757 mg) to a solution of the
amine (1 mL; 10.97 mmol) in 2% HCl (20 mL). After
2
brown-orange residue obtained after filtration was dried
to give the dye as its sodium salt with inorganic salts. A
sample (208 mg) was treated by soxhlet extraction (200
mL; 95% ethanol) to give 159 mg of a yellow dark solid.
A sample was then purified by preparative TLC (1-
completion of diazotisation (ꢁ
solution (ꢁ11 mL; 5.48 mmol) was then added drop-
wise to a solution of 4-amino benzoic acid (7) (1 equiv.;
50 mg) in 5% Na CO (50 mL). After 2 days stirring at
/
2 h), the diazonium
/
9
2
3
butanol:ethanol:ammonium
4
hydroxide:pyridine
1
/
:1:2:3) to give the pure compound. H-NMR (300
r.t., the reaction was stopped by adding 10% NaCl
solution (100 mL) and the reaction mixture filtered to
give 2.257 g of the dried dye with inorganic salts (156%
6
MHz, DMSO-d ) d 7.63 (d, 2H, Jꢂ
/
8.5 Hz, H-3 and
6.8 Hz, H-8), 7.44 (d, 1H, Jꢂ6.8
Hz, H-11), 7.40 (td, 3H, Jꢂ2.6 and 8.5 Hz, H-1, H-2
and H-6), 7.17 (t, 1H, Jꢂ6.8 Hz, H-12); HR-ES -MS
Calc. for C H N O SNa [M-Na ] 276.0443 Found:
H
1
6
H-5), 7.46 (d, 1H, Jꢂ
/
/
yield). H-NMR (300 MHz, DMSO-d ) d 7.93 (d, 2H,
H
/
Jꢂ
H-6, H-8 and H-11), 7.22 (t, 1H, Jꢂ
NMR (75 MHz, CD OD) dC 168.1(CO Na); C-IV:
/
8.3 Hz, H-3 and H-5), 7.53ꢀ
/
7.32 (m, 5H, H-1, H-2,
ꢁ
13
/
/7 Hz, H-12); C-
ꢃ ꢁ
1
2
10
3
3
3
2
2
76.0443.
148.21, 146.55, 126.16; ring: 131.85, 130.17, 127.31,
ꢁ
125.47, 120.22, 118.21 and 116.1. HR-ES -MS Calc.
for C H N O Na [M-Na ]
ꢃ ꢁ
240.0773 Found:
1
3
3
2
10
240.0773.
6.4. bis-(4-(1-Amino-4-sulfo-2-phenylazo)-phenyl)-
sulfone, di-sodium salt (14)
Dapsone (9) was tetrazotised by adding 2 mL of 4 N
NaNO to the diamine (1 g; 4.027 mmol) in 2% HCl (35
2
mL). The solution of the diazonium salt (2.013 mmol)
was added dropwise at 0 8C to a solution of sulfanilic
Fig. 5. Example of NMR numbering.

576
H. Rudyk et al. / European Journal of Medicinal Chemistry 38 (2003) 567ꢀ579
/
acid (3) (697 mg; 2 equiv.) in 5% Na CO (60 mL). After
2
xide:pyridine / 4:1:2:3) to yield two fractions of 16,
respectively 35 and 16 mg, the latter being of a lesser
3
2
detected by low resolution [ES -MS (M-2Na )
h at r.t., only a small amount of the dye could be
ꢁ
ꢃ ꢁ
1
6
degree of purity. H-NMR (300 MHz, DMSO-d ) d
H
ꢃ 2ꢁ
6
16.63; 307.2 ((M-2Na ) ]. The crude mixture was
8.73 (d, 2H, Jꢂ
8 and H-8?), 8.55 (d, 2H, Jꢂ
8.31 (s, 2H, H-2 and H-2?), 8.28 (d, 2H, Jꢂ
and H-4?), 8.10 (s, NH ), 8.08 (d, 2H, Jꢂ
/
8.3 Hz, H-14 and H-14?), 8.62 (s, 2H, H-
8.3 Hz, H-11 and H-11?),
8.1 Hz, H-4
triturated with small pieces of cotton [43] clothes and
heated at reflux for 3 h. The remaining solution was
concentrated and the water was removed by evaporation
/
/
/8.1 Hz, H-6
2
1
to give the 1.04 g dye (14). H-NMR (300 MHz, DMSO-
6
and H-6?), 7.77 (t, 2H, Jꢂ
/
8.1 Hz, H-5 and H-5?), 7.61
d ): d 7.93 (dd, 2H, Jꢂ
/
2.7 and 8.7 Hz, H-12 and H-
(t, 2H, Jꢂ
/
8.3 Hz H-13 and H-13?), 7.49 (t, 2H, Jꢂ8.3
/
H
1
3
6
1
2?), 7.86 (d, 4H, Jꢂ
/
8.8 Hz, H-3, H-3?, H-5 and H-5?),
Hz, H-14 and H-14?); C-NMR (75 MHz, DMSO-d )
dC C-IV: 153.56, 146.95, 142.68, 132.69, 132.36, 129.04,
124.53; ring: 131.11, 128.95, 128.46, 127.76, 127.59,
7
.60ꢀ
/
7.56 (m, 8H, H-2, H-2?, H-6, H-6?, H-8, H-8?, H-11
13 6
and H-11?); C-NMR (75 MHz, DMSO-d ) dC C-IV:
ꢁ
1
47.26, 146.77, 145.47, 134.87, 134.59; ring: 128.49,
28.32, 128.25, 127.89, 126.10, 118.19, 117.91; HR-
125.40, 124.46, 120.62, 118.37. HR-ES -MS Calc. for
ꢃ 2ꢁ
1
ES -MS Calc. for C H N O S Na [M-2Na ]
C H N O S Na
2
[M-2Na ]
357.0331 Found:
3
2
22
6
8
3
ꢁ
ꢃ 2ꢁ
357.0334.
2
4
18
6
8
3
2
3
07.0174 Found: 307.0174.
6.7. 4,4?-Diamino-3,3?-(3,3?-dimethoxy-biphenyl-4,4?-
6.5. bis-(4-(1-Amino-4-sulfo-2-naphtylazo)-phenyl)-
sulfone, di-sodium salt (15) [29]
diyl-bis-azo)-bis-naphthalene-1-sulfonic acid, di-sodium
salt (17) [30]
A solution of the diazonium salt of Dapsone (9)
3,3?-Dimethoxybenzidine (11) was purchased as its
(
of 4-amino-naphthalenesulfonic acid sodium salt (5) (as
2.013 mmol) was added at 0 8C dropwise to a solution
free base; a 4 N NaNO solution (2 equiv.; 1.2 mL) was
2
added at 0 8C under vigorous stirring to a solution of the
diamine (576 mg; 2.36 mmol) in 2% HCl solution. The
flask was protected from the light, as the starting
material is light sensitive. The evolution of the tetra-
zotisation was followed by MS and the reaction went to
completion after 3 h. The light brown-orange solution of
diazonium salt (2.36 mmol) was added dropwise at 0 8C
through a canular to a solution of 4-amino naphthalene
sulfonic acid sodium salt (5) (1.156 g; 2 equiv.) in
Na CO (30 mL). The evolution of the reaction was
its bis-hydrate) (987 mg; 2 equiv.) in 5% Na CO (60
3
2
mL). 2.74 g of dye was obtained after filtration as its
sodium salt with inorganic salts. A sample (345 mg) of
the crude mixture was purified by ion exchange chro-
matography to give 83 mg of the dye as its free acid
form. A sample (47 mg) was also purified using by
preparative TLC (1-butanol:ethanol:ammonium hydro-
xide:pyridine / 4:1:2:3) to give 24 mg of the pure dye
1
(
15). H-NMR (300 MHz, CD OD) d 8.75 (d, 2H, Jꢂ
/
3
H
2
3
8
8
Jꢂ
4
.1 Hz, H-14 and H-14?), 8.55 (s, 2H, H-8 and H-8?),
.32 (d, 2H, Jꢂ8.4 Hz, H-11 and H-11?), 7.89 (AB, 8H,
8.7 Hz, H-1, H-1?, H-2, H-2?, H-3, H-3?, H-4 and H-
?); 7.65 (ddd, 2H, Jꢂ1.2, 8.4 Hz, H-13 and H-13?), 7.54
ddd, 2H, H-12 and H-12?); C-NMR (75 MHz,
CD OD) d 154.82 (C-4), 150.80 (C-1), 144.82 (C-10),
monitored by MS and no coupling was observed after 2
h. The reaction mixture was stirred overnight at r.t., and
then heated for 6 h at 50 8C for completion of the
coupling. The reaction mixture was cooled down and a
20% NaCl solution (100 mL) was then added; after
filtration, the dye (17) was obtained as a dark red solid
(1.64 g) with inorganic salts. A sample (200 mg) was
treated by soxhlet extraction (200 mL; 95% ethanol) to
/
/
/
1
3
(
3
C
1
39.54, 132.56, 130.34, 130.23, 130.15 (C ), 129.69,
IV
1
ES -MS Calc. for C H N O S Na [M-2Na ]
28.18, 126.55, 125.49, 124.47 (ring) 123.51 (C-8). HR-
ꢃ 2ꢁ
ꢁ
1
give 126 mg as the pure compound 17. H-NMR (300
3
2
22
6
8
3
2
3
57.0331 Found: 357.0333.
MHz, CD OD) dH 8.74 (d, 2H, Jꢂ
1
/
8 Hz, H-14 and H-
4?); 8.61 (s, 2H, H-8 and H-8?); 8.27 (d, 2H, Jꢂ8.2 Hz,
H-5 and H-5?); 7.62ꢀ7.35 (m, 8H, H-2, H-2?, H-6, H-6?,
H-12, H-12?, H-13 and H-13?); 4.13 (s, 6H, OCH ); HR-
3
/
6.6. bis-(3-(1-Amino-4-sulfo-2-naphtylazo)-phenyl)-
sulfone, di-sodium salt (16)
/
3
ꢁ
ꢃ 2ꢁ
ES -MS Calc. for C H N O S Na [M-2Na ]
355.0627 Found: 355.0628.
3
4
26
6
8
2
2
3
by adding 4 mL of 4 N NaNO to the diamine (2 g; 8.05
,3?-Diamino biphenyl sulfone (10) was tetrazotised
2
mmol) in 2% HCl (50 mL). The solution of diazonium
salt (2.685 mmol) was added dropwise to a solution of 5
6.8. 4,4?-Diamino-3,3?-(3,3?-dimethoxy-biphenyl-4,4?-
diyl-bis-azo)-bis-naphthalene-1-sulfonic acid, di-sodium
salt (18) [31]
(
1.316 g; 2 equiv.) in 5% Na CO (50 mL). The reaction
2 3
mixture was stirred for 2 days at r.t., then a 10% NaCl
solution (30 mL) was added and 3.257 g of dye was
obtained after filtration as its sodium salt with inorganic
salts. A sample (70 mg) was purified by preparative
chromatography (1-butanol:ethanol:ammonium hydro-
3,3?-Dichlorobenzidine (13) was purchased as its
chlorhydrate salt and was delivered in isopac bottle
containing 990ꢀ/999 mg of material. 2% HCl (50 mL)
was added. A second aliquot of 30 mL was then added

H. Rudyk et al. / European Journal of Medicinal Chemistry 38 (2003) 567ꢀ
/
579
577
due to poor solubility. The solution was transferred
through a canular in a flask and a 4 N NaNO solution
(2 equiv.; 1.55 mL) was added at 0 8C under vigorous
stirring. The evolution of the tetrazotisation was fol-
6.10. 4,4?-Dihydroxy-3,3?-(3,3?-dimethoxy-biphenyl-4,4?-
diyl-bis-azo)-bis-naphthalene-1-sulfonic acid, di-sodium
salt (21) [32]
2
lowed by MS and the reaction went to completion after
2
3,3?-Dimethoxybenzidine (11) was tetrazotised by
h. A solution of diazonium salt (1.022 mmol) was
adding 4 N NaNO solution (2 equiv.; 0.65 mL) to a
2
added dropwise at 0 8C to a solution of 4-amino
naphthalene sulfonic acid, sodium salt (5) (501 mg; 2
equiv.) in Na CO solution (30 mL). The reaction
mixture was stirred overnight at r.t., after which a
sample was analysed by MS indicating the completion
of the coupling. A 20% NaCl solution (50 mL) was then
added and after filtration, the dye was obtained as a
dark red solid (586 mg) with inorganic salts. A sample
solution of the diamine (300 mg; 1.228 mmol) in 2% HCl
solution (20 mL). The light brown-orange solution of
diazonium salt (0.409 mmol) was then added dropwise
through a canular to a solution of 4-hydroxy naphtha-
lene sulfonic acid (20) (70% pure; 288 mg; 2 equiv.) in
Na CO (30 mL). The evolution of the reaction was
2
3
2
3
monitored by MS and no coupling was observed after 2
h. The reaction mixture was stirred overnight at r.t.,
after which a sample was analysed by MS indicating the
completion of the coupling. A 10% NaCl solution (100
mL) was then added and after filtration, the dye (21)
was obtained as a dark red solid (424 mg) with inorganic
salts. A sample (209 mg) was treated by soxhlet
extraction (200 mL; 95% ethanol) to give 126 mg as
(
l:ethanol:ammonium hydroxide:pyridine / 4:1:2:3) to
70 mg) was purified by preparative TLC (1-butano-
1
give 66 mg of the pure dye 18. H-NMR (300 MHz,
CD OD) d 8.91 (broad s, NH ), 8.76 (d, 2H, Jꢂ
/
8.3
8.3 Hz, H-11 and
H-11?), 8.33 (s, 2H, H-8 and H-8?), 8.15 (d, 2H, Jꢂ1.9
Hz, H-2 and H-2?), 8.11 (d, 2H, Jꢂ8.7 Hz, H-5 and H-
1.9 Hz and 8.7 Hz, H-6 and H-6?),
8.5 Hz, H-13 and H-13?), 7.52 (t, 2H,
3
H
2
Hz, H-14 and H-14?), 8.53 (d, 2H, Jꢂ
/
/
1
6
/
the pure compound 21. H-NMR (300 MHz, DMSO-d )
8.51 (d, 2H, Jꢂ8 Hz, H-14 and H-14?), 8.43 (d, 2H, Jꢂ
7.9 Hz, H-11 and H-11?), 8.16 (s, 2H, H-8 and H-8?)
7.72ꢀ7.84 (m, 2H, H-5 and H-5?), 7.50ꢀ7.32 (m, 6H, H-
2, H-2?, H-6, H-6?, H-13 and H-13?), 7.24ꢀ7.32 (m, 2H,
H-12 and H-12?); 4.12 (s, 6H, OCH ); HR-ES -MS
5
7
?), 7.93 (dd, 2H, Jꢂ
.64 (t, 2H, Jꢂ
8.5 Hz H-12 and H-12?); C-NMR (75 MHz,
/
/
/
/
1
3
Jꢂ
/
/
/
6
DMSO-d ) d C-IV: 147.51, 143.93, 139.11, 132.77,
/
C
ꢁ
1
1
32.27, 131.50, 129.15; ring: 128.66, 127.95, 126.10,
ꢁ
3
ꢃ 2ꢁ
24.91, 123.98, 122.87, 122.80, 117.50. HR-ES -MS
ꢃ 2ꢁ
Calc. for C H N O S Na [M-2Na ]
Found: 356.0467.
356.0467
3
4
24
4
10
2
2
Calc. for C H N O S Cl Na [M-2Na ]
2
359.0132
3
2
20
6
6
2
2
Found: 359.0133.
6.11. 4,4?-Dihydroxy-3,3?-(biphenyl-4,4?-diyl-bis-azo)-
bis-naphthalene-1-sulfonic acid, di-sodium salt (22) [33]
6
.9. 4,4?-Di-(2,2,2-trifluoroacetyl)-amino-3,3?-
biphenyl-4,4?-diyl-bis-azo)-bis-naphthalene-1-sulfonic
acid, di-sodium salt (19)
(
Benzidine (11) [Caution: benzidine is carcinogenic]
was tetrazotised at 0ꢀ5 8C, by adding a 4 N NaNO2
/
solution (2.05 equiv.; 0.39 mL) to a solution of the
diamine (140 mg; 0.76 mmol) in HCl N (5 mL). The
evolution of the tetrazotisation was followed by MS and
the reaction went to completion after 90 min. The
diazonium solution was then added dropwise through a
canular at 0 8C to a solution of 4-hydroxy naphthalene
sulfonic acid (20) (70% pure; 535 mg; 2 equiv.) in
saturated Na CO (10 mL). The reaction mixture was
A large excess of trifluoroacetic anhydride (0.5 mL)
was added at 0 8C to a solution of Congo red (1) (100
mg; 0.143 mL) in dry DMF (2.5 mL). The reaction
mixture was stirred and monitored by MS. The reaction
went to completion after 1 h. The solution was washed
several times with MeOH to eliminate the excess of
trifluoroacetic anhydride and the solvents were removed
under vacuum to give 124 mg of the desired compound
2
3
stirred overnight at r.t., after which a sample was
analysed by MS indicating the completion of the
coupling. A 20% NaCl solution (50 mL) was then added
and after filtration, the dye was obtained as a dark red
solid (424 mg) with inorganic salts. A sample (223 mg)
was treated by soxhlet extraction (200 mL; 95% ethanol)
1
as a light brown-orange material. H-NMR (300 MHz,
CD OD) d 9.02 (dd, 2H, Jꢂ
/
8.4 and 2.1 Hz, H-14 and
H-14?), 8.79 (s, 2H, H-8 and H-8?), 8.22 (dd, 2H, Jꢂ8.4
and 2.1 Hz, H-11 and H-11?), 8.14 (d, 4H, Jꢂ8.5 Hz, H-
8.5 Hz, H-2, H-
1.8 Hz, 6.6 and 8.6
1.8, 6.6 and 8.6
Hz, H-12 and H-12?); C-NMR (75 MHz, CD OD) dC
3
H
/
/
5
2
, H-5?, H-3 and H-3?), 8.03 (d, 4H, Jꢂ
/
1
?, H-6 and H-6?), 7.80 (ddd, 2H, Jꢂ
/
to give 188 mg as the compound 22. H-NMR (300
6
Hz, H-13 and H-13?), 7.77 (ddd, 2H, Jꢂ
/
MHz, DMSO-d ) d 8.51 (d, 2H, Jꢂ
/
8 Hz, H-14 and H-
H
1
3
14?), 8.43 (d, 2H, Jꢂ
/
7.9 Hz, H-11 and H-11?), 8.16 (s,
3
1
1
52.81, 143.52, 143.37, 142.70, 133.71, 131.38, 131.18,
2H, H-8 and H-8?) 7.80ꢀ
7.54ꢀ7.38 (m, 6H, H-2, H-2?, H-6, H-6?, H-13 and H-
13?), 7.24ꢀ7.32 (m, 2H, H-12 and H-12?); HR-ES -MS
/
7.72 (m, 2H, H-5 and H-5?),
1
28.82, 128.16, 127.86, 123.91, 123.75, 114.51. F-
9
/
ꢁ
ꢁ
NMR
C H N O S F Na [M-2Na ]
4
d ꢁ
/
73.54.
HR-ES -MS
ꢃ 2ꢁ
Calc.
421.0344 Found:
for
/
F
ꢃ 2ꢁ
Calc. for C H N O S Na [M-2Na ]
Found: 326.0364.
326.0362
3
6
20
6
8
2
6
2
32 20
4
8
3
2
21.0348.

578
H. Rudyk et al. / European Journal of Medicinal Chemistry 38 (2003) 567ꢀ579
/
6.12. 4,4?-Dihydroxy-3,3?-(3,3?-dimethoxy-biphenyl-4,4?-
diyl-bis-azo)-bis-naphthalene-6-amino-2-sulfonic acid, di-
6.14. 6,6?-Dihydroxy-3,3?-(3,3?-sulfonyl-bis-
biphenylazo)-di-benzoic acid, di-sodium salt (27)
sodium salt (24)
A solution of the diazonium salt of 3,3?-diaminodi-
phenyl sulfone (10) (2.215 mmol) was added dropwise at
0 8C to a solution of salicylic acid (25) (612 mg; 2 equiv.)
in saturated Na CO (50 mL). The reaction mixture was
3
adding 4 N NaNO solution (2 equiv.; 0.65 mL) to a
,3?-Dimethoxybenzidine (12) was tetrazotised by
2
solution of the diamine (300 mg; 1.228 mmol) in 2% HCl
solution (20 mL). The light brown-orange solution of
diazonium salt (0.409 mmol) was then added dropwise
through a canular to a solution of 6-amino-1-naphtol-3-
sulfonic acid (23) (90% pure; 217 mg; 2 equiv.) in
Na CO (30 mL). The evolution of the reaction was
2
3
stirred overnight at r.t., then heated for 2 h at 50 8C.
After addition of 10% NaCl solution to the cold reaction
mixture and filtration, 1.376 g of the dye (26) was
obtained as its sodium salt with inorganic salts. A
sample (212 mg) was purified soxhlet extraction to give
2
3
1
6
monitored by MS and no coupling was observed after 2
h. The reaction mixture was stirred overnight at r.t.,
after which a sample was analysed by MS indicating the
reaction was not complete, therefore the reaction
mixture was heated first 2 h at 50 8C then 2 days
further; the advancement of the reaction was monitored
and after 2 days, as no further evolution could be
observed a 10% NaCl solution (50 mL) was then added
and after filtration, the dye (24) was obtained as a dark
red solid (901 mg) with inorganic salts and remaining
starting material (23). A sample (211 mg) was treated by
soxhlet extraction (200 mL; 95% ethanol) to give 149 mg
207 mg of 26. H-NMR (300 MHz, DMSO-d ) d 8.29
H
(d, H, Jꢂ
Hz, H-2 and H-2?), 8.06 (dd, 4H, Jꢂ
4, H-4?, H-6 and H-6?), 7.82 (dd, 2H, Jꢂ
H-10 and H-10?), 7.77 (t, 2H, Jꢂ7.8 Hz, H-5 and H-5?),
6.73 (d, 1H, Jꢂ8.8 Hz, H-11 and H-11?); HR-ES -MS
Calc. for C H N O SNa [M-2Na ]
/
2.8 Hz, H-8 and H-8?), 8.28 (d, 2H, Jꢂ
2.1 and 7.8 Hz, H-
2.8 and 8.8 Hz,
/
2.1
/
/
/
ꢁ
/
ꢃ 2ꢁ
272.0345
2
6
16
4
8
2
Found: 272.0342.
6.15. 6,6?-Dihydroxy-3,3?-biphenyl-4,4?-diyl-bis-azo-di-
benzoic acid, di-sodium salt (28) [37,38]
1
as the pure compound 24. H-NMR (300 MHz, DMSO-
Benzidine was tetrazotised at 0ꢀ5 8C, by adding a 4 N
/
6
d ) d 7.98 (d, 2H, Jꢂ
/
8.6 Hz, H-5 and H-5?), 7.93 (d,
8.5 Hz, H-10 and H-10?), 7.42 (d, 2H, Jꢂ8.6
Hz, H-6 and H-6?), 7.42 (s, 2H, H-14 and H-14?), 7.22 (s,
NaNO solution (2.05 equiv.; 0.39 mL) to a solution of
H
2
2
H, Jꢂ
/
/
the diamine (140 mg; 0.76 mmol) in HCl N (5 mL). The
reaction mixture was stirred for 90 min. The diazonium
solution was then added to a solution of salicylic acid
(25) (2 equiv.) in saturated Na CO solution (10 mL). A
2
1
4
H, H-2 and H-2?), 6.65 (d, 2H, Jꢂ8.4 Hz, H-11 and H-
/
1?), 6.63 (s, 2H, H-13 and H-13?), 6.41 (broad s, NH ),
2
2
3
ꢁ
.07 (s, 6H, OCH3); HR-ES -MS Calc. for
ꢃ 2ꢁ
10% NaCl solution (30 mL) was then added. The dye
was filtered and dried overnight. A sample was purified
by preparative chromatography to give the pure com-
C H N O S Na [M-2Na ]
3
371.0576 Found:
4
26
6
10
2
2
3
71.0577.
1
pound 28. H-NMR (300 MHz, CD OD) d 8.29 (d, H,
3
H
6
.13. 6,6?-Dihydroxy-3,3?-(4,4?-sulfonyl-bis-
biphenylazo)-di-benzoic acid, di-sodium salt (26) [39]
Jꢂ
3, H-3?, H-4, H-4?, H-6 and H-6?), 7.81 (dd, 2H, Jꢂ
and 8.7 Hz, H-10 and H-10?), 6.77 (d, 2H, Jꢂ8.7 Hz, H-
11 and H-11?); HR-ES -MS Calc. for C H N O Na
/
2.6 Hz, H-8 and H-8?), 7.29 (AB, 8H, H-2, H-2?, H-
/2.6
/
ꢁ
A solution of the diazonium salt of 4,4?-diaminodi-
phenylsulfone (9) (2.01 mmol) was added dropwise to a
solution of salicylic acid (25) (556 mg; 2 equiv.) in 5%
Na CO (50 mL). The reaction mixture was stirred
overnight at r.t., then heated for 6 h at 50 8C. Addition
of 10% NaCl solution to the cold mixture and ensuing
filtration gave 1.82 g (153%) of dye as a brown solid as
its sodium salt with inorganic salts. A sample (214 mg)
was treated by soxhlet extraction (200 mL; 95% ethanol)
2
6
16
4
6
2
ꢃ 2ꢁ
[M-2Na ]
240.0535 Found: 240.0531.
6.16. 6,6?-Dihydroxy-3,3?-(3,3?-dimethoxy-biphenyl-4,4?-
diyl-bis-azo)-dibenzoic acid, di-sodium salt (29)
2
3
3,3?-Dimethoxybenzidine (12) was tetrazotised by
adding 4 N NaNO solution (2 equiv.; 0.95 mL) to a
2
solution of the diamine (519 mg; 2.124 mmol) in 2% HCl
solution (30 mL). The light brown-orange solution of
diazonium salt was then added dropwise through a
canular to a solution of salicylic acid (25) (587 mg; 2
equiv.) in saturated Na CO (30 mL). The evolution of
1
to give 184 mg as the compound 26. H-NMR (300
MHz, CD OD) d 8.53 (d, 2H, Jꢂ
/
2.5 Hz, H-8 and H-
8.6 Hz, H-2, H-2?, H-3, H-3?, H-5,
H-5?, H-6 and H-6?), 7.96 (dd, 2H, Jꢂ2.5, 8.8 Hz, H-10
and H-10?), 6.95 (d, 2H, Jꢂ8.8 Hz, H-11 and H-11?);
C-NMR (75 MHz, CD OD) dC C-IV: 173.46
3
C
8
?), 8.09 (AB, 8H, Jꢂ
/
/
2
3
/
the reaction was monitored by MS the reaction mixture
was kept overnight under stirring at r.t. A 10% NaCl
solution (150 mL) was then added and after filtration,
the dye was obtained as a solid (869 mg) with inorganic
salts. A sample (221 mg) was treated by soxhlet
extraction (200 mL; 95% ethanol) to give 180 mg as
1
3
3
(
CO Na), 166.13, 155.38, 144.53, 141.64, 118.99; ring:
2
ꢁ
128.51, 127.91, 126.13, 122.69, 116.95; HR-ES -MS
ꢃ 2ꢁ
Calc. for C H N O SNa [M-2Na ]
272.0345
2
6
16
4
8
2
Found: 272.0343.

H. Rudyk et al. / European Journal of Medicinal Chemistry 38 (2003) 567ꢀ
/
579
579
1
[
[
[
[
[
15] S.J. Pollack, I.I. Sadler, S.R. Hawtin, V.J. Tailor, M.S. Shearman,
Neurosci. Lett. 197 (1995) 211ꢀ214.
16] S. Caspi, M. Halimi, A. Yanai, S. BenSasson, A. Taraboulos, R.
Gabizon, J. Biol. Chem. 273 (1998) 3484ꢀ3489.
17] L. Ingrosso, A. Ladogana, M. Pocchiari, J. Virol. 69 (1995) 506ꢀ
the pure compound 29. H-NMR (300 MHz, d6-
DMSO) dH 8.26 (d, 2H, Jꢂ2.6 Hz, H-8 and H-8?),
2.6 and 8.7 Hz, H-10 and H-10?),
7.54 (m, 2H, H-5 and H-5?), 7.47ꢀ7.40 (m, 2H, H-
and H-6?), 7.36ꢀ7.28 (m, 2H, H-2 and H-2?), 6.79 (dd,
H, Jꢂ2.5 and 8.7 Hz, H-11 and H-11?), 4.09
/
/
7
7
6
2
.82ꢀ
/
7.77 (m, 2H, Jꢂ
/
/
.65ꢀ
/
/
/
/
508.
18] L.E. Gray, R.J. Kavlock, J. Ostby, J. Ferrell, Teratology 27
(1983) A46.
19] L.E. Gray, J. Ostby, R. Kavlock, R. Marshall, Fundam. Appl.
/
ꢁ
(
s, 3H,OMe);HR-ES -MS Calc. for C H N O Na
28 20 4 8 2
ꢃ 2ꢁ
[
M-2Na ]
270.0641 Found: 270.0640.
Toxicol. 19 (1992) 411ꢀ
/422.
[
[
20] L.E. Gray, J. Ostby, Fundam. Appl. Toxicol. 20 (1993) 177ꢀ183.
21] R.P. Bos, J.P. Koopman, J.L.G. Theuws, P.T. Henderson, Mutat.
Res. 181 (1987) 327.
/
Acknowledgements
[22] T. Zenser, V. Lakshmi, B. Davis, Drug Metab. Dispos. 26 (1998)
8
56ꢀ
[23] R. Demaimay, J. Harper, J. Gordon, D. Weaver, B. Chesebro, B.
Caughey, J. Neurochem. 71 (1998) 2534ꢀ2541.
/859.
We would like to acknowledge the MRC for financial
support (HR, SV), the EPSRC Mass Spectrometry
Service Centre at Swansea for accurate mass spectra
and Cardiff Centre for Molecular Modelling for use of
the computing facilities.
/
[
24] H. Zollinger, Color Chemistry, Synthesis, Properties and Appli-
cation of Organic Dyes and Pigments, Weinheim VCH, 1987.
25] Bad. Anilin-u. Sodaf., DE 84893, Fortschr. Teerfarbenfabr. Verw.
Industriezweige, DE, GE, 4, 847.
[
[
26] J. Szadowski, Dyes Pigments 14 (1990) 217ꢀ
27] H.S. Freeman, S.-D. Kim, R.D. Gilbert, R. Mcgregor, Dyes
Pigments 17 (1991) 83ꢀ100.
28] H. Rudyk, S. Vasiljevic, C.R. Birkett, J. Hope, I.H. Gilbert, J.
Gen. Virol. 81 (2000) 1155ꢀ1164.
[29] P.P.T. Sah, Recl. Trav. Chim. Pays-Bas 69 (1950) 1407ꢀ
[30] Bayer Co., DE 38802; Fortschr. Teerfarbenfabr. Verw.
Industriezweige; DE, GE, 1, 488.
[31] F.H. Holmes, Trans. Faraday Soc. 54 (1958) 1166ꢀ
[32] Bayer Co.; DE 38802; Fortschr. Teerfarbenfabr. Verw.
Industriezweige; DE, GE, 1, 489, 491.
/
224.
[
/
References
[
/
[
1] I.H. Gilbert, H. Rudyk, Int. Antiviral News 7 (1999) 78ꢀ
2] S.B. Prusiner, M.R. Scott, S.J. DeArmond, F.E. Cohen, Cell 93
1998) 337ꢀ348.
3] S. Leimann, R. Glockshuber, Biochem. Biophys. Res. Commun.
50 (1998) 187ꢀ193.
/
81.
/
1434.
[
&
(
/
[
[
/
1171.
2
/
&
4] G.A.H. Wells, A.C. Scott, C.T. Johnson, R.F. Gunning, R.D.
Hancock, M. Jeffrey, M. Dawson, R. Bradley, Vet. Rec. 121
[33] A. Arcoria, G. Scarlata, Ann. Chim. (Rome) 57 (1967) 1125.
[34] H. Zollinger, Azo and Diazo Chemistry, Aliphatic and Aromatic
Compounds, Interscience Publishers, Inc., New York, 1961.
[35] K.H. Saunders, R.L.M. Allen, Aromatic Diazo Compounds,
third ed., Edward Arnold Publishers Ltd, London, 1985.
(
1987) 419ꢀ
5] D.R. Brown, A. Besinger, J. Biochem. 334 (1998) 423ꢀ
6] N. Nishida, Cell. Mol. Neurobiol. 17 (1997) 537ꢀ545.
/
420.
[
[
[
/429.
/
7] I. Tobler, S.E. Gaus, T. Deboer, P. Achermann, M. Fischer, T.
Rulicke, M. Moser, B. Oesch, P.A. McBride, J.C. Manson,
[36] C.A. Mathis, K. Mahmood, M.L. Debnath, W. Klunk, J.
Nature 380 (1996) 639ꢀ
8] J.W. Herms, A. Madlung, D.R. Brown, H.A. Kretzschmar, Glia
1 (1997) 253ꢀ257.
9] K. Basler, B. Oesch, M. Scott, D. Westaway, M. Walchli, D.F.
Groth, M.P. McKinley, S.B. Prusiner, C. Weissmann, Cell 46
/
642.
Labelled Compd. Radiopharm. 40 (1997) 94ꢀ
[37] K. Brass, W. Wittenberger, Ber. Deut. Chem. Gesellschaft 68
(1935) 1905ꢀ1912.
[38] Bayer Co., DE 31658, Fortschr. Teerfarbenfabr. Verw.
/
95.
[
[
2
/
/
&
Industriezweige; DE, GE, 1, 465.
(
1986) 417ꢀ
10] B.W. Caughey, A. Dong, K.S. Bhat, D. Ernst, S.F. Hayes, W.S.
Caughey, Biochemistry 31 (1991) 7672ꢀ7680.
11] B. Caughey, R.E. Race, D. Ernst, M.J. Buchmeier, B. Chesebro,
J. Virol. 63 (1989) 175ꢀ181.
12] W.E. Klunk, M.L. Debnath, J.W. Pettegrew, Neurobiol. Aging 15
1994) 691ꢀ698.
13] B. Caughey, G.J. Raymond, J. Virol. 67 (1993) 643ꢀ
14] T. Bandiera, J. Lanson, C. Post, M. Varasi, Curr. Med. Chem. 4
1997) 159ꢀ170.
/
428.
[39] T. Dewing, W.H. Gray, B.C. Platt, D. Stephenson, J. Chem. Soc.
[
[
[
(1942) 239ꢀ244.
/
/
[40] M. Skowronek, B. Stopa, L. Konieczny, J. Rybarska, B.
Piekarska, E. Snezler, G. Bakalarski, I. Roterman, Biopolymers
/
46 (1998) 267ꢀ
[41] B. Stopa, L. Konieczny, B. Piekarska, I. Roterman, J. Rybarska,
M. Skowronek, Biochimie 79 (1997) 23ꢀ26.
[42] B. Stopa, M. G o´ rny, L. Konieczny, B. Piekarska, J. Rybarska, M.
Skowronek, I. Roterman, Biochimie 80 (1998) 963ꢀ968.
[43] J. Szadowski, Dyes Pigments 14 (1990) 217.
/
281.
(
/
/
[
[
/
650.
/
(
/